""" Boiler plate
- import all packages
- emulate snakemake so can load config as usual
- load the network you want
TODO: make it easier to restore old runs
TODO: hide all the boiler pates in a file
"""
%load_ext autoreload
%autoreload 2
%precision %e
import logging
import pypsa
import os.path

import cartopy.crs as ccrs
import matplotlib.pyplot as plt
import pandas as pd
# running the jupyter notebook on the compute nodes doesnt build the path as expected, you have to manually do this
import IPython

working_directory = os.path.dirname(IPython.extract_module_locals()[1]['__vsc_ipynb_file__'])
workflow_dir = os.path.dirname(working_directory)
scripts_dir = os.path.join(workflow_dir, "scripts")
root_dir = os.path.dirname(workflow_dir)

os.chdir(scripts_dir)


logging.basicConfig(level=logging.CRITICAL)
logger = logging.getLogger(__name__)


from _helpers import configure_logging, mock_snakemake
# from make_summary import assign_carriers
import _plot_utilities
import plot_network 
logging.getLogger('plot_network').setLevel(logging.CRITICAL)
logging.getLogger('_plot_utilities').setLevel(logging.CRITICAL)

from constants import PLOT_COST_UNITS, PLOT_CAP_UNITS,PLOT_SUPPLY_UNITS
from _plot_utilities import fix_network_names_colors, determine_plottable
from plot_network import plot_cost_map, plot_map

The autoreload extension is already loaded. To reload it, use:
  %reload_ext autoreload

REMIND_COUPLED = True
PLANNING_YEAR = 2035
config = None if not REMIND_COUPLED else "resources/tmp/remind_coupled.yaml"

snakemake = mock_snakemake(
    "plot_network",
    snakefile_path=workflow_dir,
    topology="current+FCG",
    # co2_pathway="exp175default",
    co2_pathway="SSP2-PkBudg1000-PyPS",
    planning_horizons=PLANNING_YEAR,
    configfiles = config,
    heating_demand="positive",
)

configure_logging(snakemake, logger=logger)
config = snakemake.config
tech_colors = config["plotting"]["tech_colors"]


ntw_path = snakemake.input.network
# ntw_path = f"/home/ivanra/downloads/PaperResultsXiaowei_networks/postnetwork-ll-current+Neighbor-exponential175-{PLANNING_YEAR}.nc"
# ntw_path = "/p/tmp/ivanra/xiaowei-pypsa/PyPSA-China/results/version-0325.175.1H/postnetworks/positive/postnetwork-ll-current+Neighbor-exponential175-2060.nc"
n = pypsa.Network(ntw_path)
results_dir = os.path.dirname(os.path.dirname(ntw_path))
weighting = n.snapshot_weightings.iloc[0,0]

2025-06-06 09:50:24,057 - _helpers.py - INFO - =========== NEW RUN ===========
2025-06-06 09:50:24,057 - _helpers.py - INFO - =========== NEW RUN ===========
2025-06-06 09:50:24,057 - _helpers.py - INFO - =========== NEW RUN ===========
INFO:__main__:=========== NEW RUN ===========

Fix network for plotting // add missing info

fix_network_names_colors(n, config)
determine_plottable(n)

/p/tmp/ivanra/PyPSA-China-PIK/workflow/scripts/_plot_utilities.py:232: FutureWarning: A value is trying to be set on a copy of a DataFrame or Series through chained assignment using an inplace method.
The behavior will change in pandas 3.0. This inplace method will never work because the intermediate object on which we are setting values always behaves as a copy.

For example, when doing 'df[col].method(value, inplace=True)', try using 'df.method({col: value}, inplace=True)' or df[col] = df[col].method(value) instead, to perform the operation inplace on the original object.


  n.carriers.color.fillna(NAN_COLOR, inplace=True)

Plot maps

plot capex

from plot_network import plot_energy_map
from _plot_utilities import set_plot_style
set_plot_style(
    style_config_file= os.path.abspath("../../config/plotting_styles/network_map.mplstyle"),
    base_styles=["classic", "seaborn-v0_8-white"],
)

with io.capture_output() as captured:
    ax = plot_cost_map(n, config["plotting"], cost_pannel=True, capex_only=True)

# TODO fix layout
with io.capture_output() as captured:
    ax = plot_cost_map(n, config["plotting"], cost_pannel=True, plot_additions=False, capex_only=False)

plot electricy generation map

from IPython.utils import io

with io.capture_output() as captured:
    plot_energy_map(n,config["plotting"], carrier="AC", components = ["Generator", "Link"])

from plot_network import plot_energy_map
from _plot_utilities import set_plot_style

if snakemake.config["heat_coupling"]:
    set_plot_style(
        style_config_file= os.path.abspath("../../config/plotting_styles/network_map.mplstyle"),
        #snakemake.config["plotting"]["network_style_config_file"],
        base_styles=["classic", "seaborn-v0_8-white"],
    )
    plot_energy_map(n,config["plotting"], carrier="heat", components = ["Generator", "Link"])

Nodal prices

from plot_network import plot_nodal_prices
with io.capture_output() as captured:
    plot_nodal_prices(n, config["plotting"], "AC")

Interactive plot map

import numpy as np
# make names and link sizes, make sure we only plot AC, DC & statiosn
ac_links = n.links[n.links.carrier == "AC"]
colors = n.links.index.to_series().apply(lambda x: 'black' if 'ext' in x else 'pink')
widths = np.log(n.links.p_nom_opt + 3) / 2
widths[~widths.index.isin(ac_links.index)] = 0
widths[widths.index.str.contains('reversed')] = 0
names = n.links.copy()
names["name"] = names.index.values
names.loc[~names.index.isin(ac_links.index), "p_nom_opt"] = ""
names.loc[~names.index.isin(ac_links.index), "name"] = ""
buses = n.buses.copy()
buses["name"] = buses.apply(lambda x: f"{x.name}" if x.carrier == "AC" or x.carrier=="stations" else "", axis=1)
buses["sizes"] = buses.apply(lambda x:10 if x.carrier == "AC" or x.carrier=="stations" else 0, axis=1)
buses["colors"] = buses.apply(lambda x: "black" if x.carrier == "AC" else "red", axis=1)
buses.carrier.unique()
ax = n.iplot(link_colors=colors, link_widths=widths, link_text = names.p_nom_opt.astype(str) + names.name, bus_text=buses.name, bus_sizes=buses.sizes, bus_colors=buses.colors)

/p/tmp/ivanra/anaconda/ipykernel_3020255/316105652.py:10: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise an error in a future version of pandas. Value '' has dtype incompatible with float64, please explicitly cast to a compatible dtype first.
  names.loc[~names.index.isin(ac_links.index), "p_nom_opt"] = ""

extendable lines

CO2 emissions & Stores

n.global_constraints

attribute	type	investment_period	carrier_attribute	sense	constant	mu
GlobalConstraint

fig, ax = plt.subplots()
# calc the t resolved CO2 emissions from generators
time_res_emissions = ((n.generators_t.p/ n.generators.efficiency)* n.generators.carrier.map(n.carriers.co2_emissions)).T.groupby(n.generators.carrier).sum()
colors = time_res_emissions.T.columns.map(tech_colors).fillna("lightgrey")
time_res_emissions.T.plot(ax =ax, color =  colors, legend = True)

<Axes: xlabel='snapshot'>

gen_emissions = ((n.generators_t.p/ n.generators.efficiency)* n.generators.carrier.map(n.carriers.co2_emissions)).T.groupby(n.generators.carrier).sum().T.cumsum()
fig, ax = plt.subplots()
gen_emissions.where(gen_emissions>=0).dropna(axis=1, how="all").plot(ax=ax)
ax.set_ylabel("cum CO2 emissions [t]")

Text(0, 0.5, 'cum CO2 emissions [t]')

try:
    fig, ax = plt.subplots()
    # calc the t resolved CO2 emissions from generators
    ((n.generators_t.p/ n.generators.efficiency)* n.generators.carrier.map(n.carriers.co2_emissions)).T.groupby(n.generators.carrier).sum().sum().T.cumsum().plot(ax =ax, label = "CO2 emissions (cum)", lw=4, c ="black", ls = "--")
    n.stores_t.e.T.groupby(n.stores.carrier).sum().loc[["CO2","H2", "CO2 capture", "gas", "biomass"]].T.plot(lw=4, cmap ="jet", ax =ax)
    ax.legend()
    ax.semilogy()
    ax.set_ylim(1e3,1e10)
    ax.set_ylabel("carrier stock")
except KeyError:
    print("Overnight model does not have CO2 capture or biomass stores, skipped")

Overnight model does not have CO2 capture or biomass stores, skipped

gas_stores = n.stores[n.stores.carrier.str.contains("gas")]
gas_stores_t = n.stores_t.e[n.stores_t.e.columns.intersection(gas_stores.index)]
gas_stores_t.sum(axis=1).plot()

<Axes: xlabel='snapshot'>

CO2 capture

! beware the store is the difference :)

try:
    stores = n.stores_t.e.T.groupby(n.stores.carrier).sum()
    diff = stores.iloc[:, -1] -stores.iloc[:, 0]
    co2_cap = stores.iloc[:, -1].loc[["CO2 capture"]].sum()
    co2_cap
except KeyError:
    print("Overnight model does not have CO2 capture or biomass stores, skipped")

Overnight model does not have CO2 capture or biomass stores, skipped

n.stores_t.e.T.groupby(n.stores.carrier).sum().T.plot(lw=4, cmap ="jet")

<Axes: xlabel='snapshot'>

Time series

Weekly energy balances

from plot_time_series import plot_energy_balance

with io.capture_output() as captured:
    ax = plot_energy_balance(n, config["plotting"], start_date=f"{PLANNING_YEAR}-01-01 01:00", end_date=f"{PLANNING_YEAR}-12-31 12:00:00")
    ax.grid(axis='y')
    ax.set_title("Electricity Balance")

with io.capture_output() as captured:
    ax = plot_energy_balance(n, config["plotting"], start_date=f"{PLANNING_YEAR}-07-24 00:00", end_date=f"{PLANNING_YEAR}-07-31 12:00:00")
    ax.grid(axis='y')
    ax.set_title("Electricity Balance")

with io.capture_output() as captured:
    ax = plot_energy_balance(n, config["plotting"], start_date=f"{PLANNING_YEAR}-03-31 21:00", end_date=f"{PLANNING_YEAR}-04-06 12:00:00")
    ax.grid(axis='y')
    ax.set_title("Electricity Balance")

price series

from constants import PROV_NAMES 
marginal_price_series = n.buses_t["marginal_price"][PROV_NAMES]
prov_max_price_series = marginal_price_series.T.max()
prov_min_price_series = marginal_price_series.T.min()
spread = marginal_price_series.T.max()-marginal_price_series.T.min()
load_ac = n.statistics.withdrawal(bus_carrier="AC", aggregate_time=False, groupby="location", comps = "Load")

regional_weighed_price = marginal_price_series.T.mul(load_ac).T.sum()/load_ac.T.sum()
order =regional_weighed_price.sort_values().index

weighed_prices = (marginal_price_series.T.mul(load_ac)/load_ac.sum()).sum()
ax = marginal_price_series[order].plot(alpha=0.5, lw=2, legend=False, cmap="plasma")
weighed_prices.plot(ax = ax, lw=3, c="black", label = "weighted price avg", alpha =0.8, ls = "--")
ax.legend(  loc='upper left', bbox_to_anchor=(1, 1), ncols=2);

from _plot_utilities import find_weeks_of_interest
from constants import PROV_NAMES

winter_week, summer_week = find_weeks_of_interest(n, f"{PLANNING_YEAR}-04-01", f"{PLANNING_YEAR}-10-06")
marginal_price_series = n.buses_t["marginal_price"][PROV_NAMES]
prov_max_price_series = marginal_price_series.T.max()
prov_min_price_series = marginal_price_series.T.min()
spread = marginal_price_series.T.max()-marginal_price_series.T.min()
load_ac = n.statistics.withdrawal(bus_carrier="AC", aggregate_time=False, groupby="location", comps = "Load")
weighed_prices = (marginal_price_series.T.mul(load_ac)/load_ac.sum()).sum()


ax = prov_min_price_series.loc[summer_week].plot(label = "Marginal price\n(least expensive province)")
prov_max_price_series.loc[summer_week].plot(label = "Marginal price\n(most expensive province)", ax =ax)
# spread.loc[summer_week].plot(label = "Price spread", ax =ax, ls= "--")
ax.set_ylabel("Price [€/MWh]")
# ax.vlines(winter_week[len(winter_week)//2], 0,prov_max_price_series.max()*1.2, color="black", ls = "--", label="peak event")
ax.vlines(summer_week[len(summer_week)//2], 0,prov_max_price_series.loc[summer_week].max()*1.2, color="black", ls ="--", label="peak event")

weighed_prices.loc[summer_week].plot(ax=ax, label ="national mean weighted price")
ax.legend(bbox_to_anchor=(1.05, 1), loc='upper left', title="Legend");

ax = prov_min_price_series.plot(label = "Marginal price\n(least expensive province)", logy=True)
ax = prov_max_price_series.plot(label = "Marginal price\n(most expensive province)", logy=True)
ax.set_ylabel("Price [€/MWh]")
ax.vlines(winter_week[len(winter_week)//2], 0,prov_max_price_series.max()*1.2, color="black", ls = "--", label="peak event")
ax.vlines(summer_week[len(summer_week)//2], 0,prov_max_price_series.max()*1.2, color="black", ls ="--", label="peak event")

weighed_prices.plot(ax=ax, label ="national mean weighted price", logy=True)
ax.legend(bbox_to_anchor=(1.05, 1), loc='upper left', title="Legend");

Prices - balance overlay

with io.capture_output() as captured:
    ax = plot_energy_balance(n, config["plotting"], start_date=str(winter_week[0]), end_date=str(winter_week[-1]))
    ax.grid(axis='y')
    ax.set_title("Electricity Balance - worst week winter")
    ax2 = ax.twinx()
    prov_max_price_series.loc[winter_week].plot(ax=ax2, color="red", lw=2, ls = "--", label= "Most exp Prov")
    weighed_prices.loc[winter_week].plot(ax=ax2, color = "blue", lw=3, ls = ":",label = "National weighted price")
    fig = ax.get_figure()
    ax2.set_ylabel("PriceEur/MWh", color="red")
    ax2.tick_params(axis='y', labelcolor='red')
    # Move the existing legend to the right
    handles, labels = ax.get_legend_handles_labels()
    handles2, labels2 = ax2.get_legend_handles_labels()
    ax.legend(handles + handles2, labels + labels2, loc='center left', bbox_to_anchor=(1.1, 0.5))

with io.capture_output() as captured:
    ax = plot_energy_balance(n, config["plotting"], start_date=str(summer_week[0]), end_date=str(summer_week[-1]))
    ax.grid(axis='y')
    ax.set_title("Electricity Balance - worst week summer")
    ax2 = ax.twinx()
    prov_max_price_series.loc[summer_week].plot(ax=ax2, color="red", lw=2, ls = "--", label= "Most exp Prov")
    weighed_prices.loc[summer_week].plot(ax=ax2, color = "blue", lw=3, ls = ":",label = "National weighted price")
    fig = ax.get_figure()
    ax2.set_ylabel("PriceEur/MWh", color="red")
    ax2.tick_params(axis='y', labelcolor='red')
    # Move the existing legend to the right
    handles, labels = ax.get_legend_handles_labels()
    handles2, labels2 = ax2.get_legend_handles_labels()
    ax.legend(handles + handles2, labels + labels2, loc='center left', bbox_to_anchor=(1.1, 0.5))

if snakemake.config["heat_coupling"]:
    ax = plot_energy_balance(n, config["plotting"], start_date=f"{PLANNING_YEAR}-03-31 21:00", end_date=f"{PLANNING_YEAR}-09-06 12:00:00", bus_carrier="heat")
    ax.set_title("Heat balance")
    ax2 = ax.twinx()
    prov_max_price_series.loc[f"{PLANNING_YEAR}-03-31 21:00":f"{PLANNING_YEAR}-09-06 12:00:00"].plot(ax=ax2, color="red", lw=2, ls = "--", label= "Most exp Prov")
    weighed_prices.loc[f"{PLANNING_YEAR}-03-31 21:00":f"{PLANNING_YEAR}-09-06 12:00:00"].plot(ax=ax2, color = "blue", lw=3, ls = ":",label = "National weighted price")
    prov_min_price_series.loc[f"{PLANNING_YEAR}-03-31 21:00":f"{PLANNING_YEAR}-09-06 12:00:00"].plot(ax=ax2, color="green", lw=2, ls = "--", label= "Least exp Prov")

    fig = ax.get_figure()
    ax2.set_ylabel("PriceEur/MWh", color="red")
    ax2.tick_params(axis='y', labelcolor='red')
    ax2.set_ylim([0, prov_max_price_series.loc[f"{PLANNING_YEAR}-03-31 21:00":f"{PLANNING_YEAR}-09-06 12:00:00"].max()*3])
    # Move the existing legend to the right
    handles, labels = ax.get_legend_handles_labels()
    handles2, labels2 = ax2.get_legend_handles_labels()
    ax.legend(handles + handles2, labels + labels2, loc='center left', bbox_to_anchor=(1.1, 0.5))

if snakemake.config["heat_coupling"]:
    ax = plot_energy_balance(n, config["plotting"], start_date=f"{PLANNING_YEAR}-01-01 00:00", end_date=f"{PLANNING_YEAR}-12-31 23:00:00", bus_carrier="heat")
    ax.set_title("Heat balance")

from plot_time_series import plot_residual_load_duration_curve, plot_load_duration_curve, plot_regional_load_durations

LOAD DURATION CURVES

fig, ax = plt.subplots(1, 2, figsize=(10, 6), sharey=True)
plot_load_duration_curve(n, carrier="AC", ax = ax[0])
plot_residual_load_duration_curve(n, ax = ax[1])
fig.tight_layout()

with io.capture_output() as captured:

    plot_regional_load_durations(n, carrier="AC", cmap = "viridis")

fix, ax = plt.subplots()
ds_AC = n.statistics.withdrawal(bus_carrier="AC", aggregate_time=False).loc[("Load", "-")]/1e3
ds_AC.plot(ax=ax, label="Electricity",  c="orange")
if snakemake.config["heat_coupling"]:
    ds_heat = n.statistics.withdrawal(bus_carrier="heat", aggregate_time=False).loc[("Load", "-")]/1e3
    ds_heat.plot(ax=ax, label="Heat", c = "blue")  
ax.legend()
ax.set_ylabel("EnergyDemand / GW")

Text(0, 0.5, 'EnergyDemand / GW')

Power flows map

from _pypsa_helpers import get_location_and_carrier
transmission = n.statistics.transmission(
        bus_carrier="AC",
        groupby=get_location_and_carrier,
        aggregate_time=False,
    )/1e6 # # convert to TW 
transm_tot = transmission.T.sum().reset_index()
transm_tot = transm_tot.set_index("location").rename(columns={0:"HV transmission"}).sort_values(by="HV transmission")
regi_transmission =  transmission.groupby("location").sum().loc[transm_tot.index[::-1]]

ax = (
    regi_transmission.T.loc[f"{PLANNING_YEAR}-07-20":f"{PLANNING_YEAR}-07-27"]
    .plot.area(stacked=True, cmap="magma_r")
)
ax.legend(loc="center left", bbox_to_anchor=(1, 0.5), title="Prov", ncols=2)
ax.set_ylabel("Transmission [TWh]");

ax = transm_tot.plot.bar(y="HV transmission")
ax.set_ylabel("HV Transmission / MWHr");

Plot cap factors time series

capacity_factors = n.statistics.capacity_factor(aggregate_time=False).loc[["Generator"]].droplevel(0).T
colors_lowered = {k.lower().rstrip():v for k,v in config["plotting"]["tech_colors"].items()}
cap_colors = {k: colors_lowered.get(k.lower().rstrip(), "lightgrey") for k in [c for c in capacity_factors.columns]}

axes = capacity_factors.dropna(axis=1, how="all").fillna(0).plot(subplots=True, figsize=(10,10), color= cap_colors)
for ax in axes:
    ax.set_ylim([0,1.1])
    ax.set_yticks([0,0.3, 0.6,1])

region_rev = n.statistics.revenue(groupby = ["location", "carrier"], bus_carrier="AC", aggregate_time=False).fillna(0)
rev_regional = region_rev[(region_rev<0).all(axis=1)].groupby("location").sum().T
prod_reg = n.statistics.energy_balance(groupby = ["location", "carrier"], bus_carrier="AC", aggregate_time=False)*weighting
consum_regional = prod_reg.query("component=='Load'").groupby("location").sum().T
balance = prod_reg.groupby("location").sum().T
supp_regional = prod_reg.query("component!='Load'").groupby("location").sum().T
ax = (rev_regional/consum_regional).plot(cmap="viridis")
ax.set_ylabel("Expenditure/Withdrawal [€/MWh]")
ax.set_xlabel("Time")
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5), ncols=2)

<matplotlib.legend.Legend at 0x7f0879bca9c0>

data = pd.concat([supp_regional.sum(), -1*consum_regional.sum(), balance.sum()], axis=1)
data.columns = ["Supply", "Consum", "Balance"]
data = data.query("index != ''") # no location assigned
fig, ax = plt.subplots(figsize=(10, 6))
bar_width = 0.25
x = np.arange(len(data.index))
data = data/1e6

ax.bar(x - bar_width, data["Supply"], width=bar_width, label="Supply")
ax.bar(x, data["Consum"], width=bar_width, label="Consum")
ax.bar(x + bar_width, data["Balance"], width=bar_width, label="Balance")

ax.set_xticks(x)
ax.set_xticklabels(data.index, rotation =70, ha ='right', fontsize=11)
ax.legend()
ax.set_ylabel("Energy [TWh]", fontsize=12)
fig.tight_layout()
plt.show()

max_val = max(supp_regional.max().max(), -1*consum_regional.max().max())

norm_price =  n.buses_t["marginal_price"][PROV_NAMES].loc[f"{PLANNING_YEAR}-07-22":f"{PLANNING_YEAR}-07-28"]
norm_price/= norm_price.max()
norm_price*=(consum_regional*-1e-6).loc[f"{PLANNING_YEAR}-07-22":f"{PLANNING_YEAR}-07-28"].max().max()*0.8
consum_regional = consum_regional[PROV_NAMES]

def plot_sufficience_panels(consum_regional, supp_regional):
    fig, axes = plt.subplots(6,6, sharex=True, sharey=True, figsize=(20,20))
    for i,c in enumerate(consum_regional.columns):
        k, l = (i+4)//6, (i+4)%6
        (consum_regional[c]*-1e-3).loc[f"{PLANNING_YEAR}-07-22":f"{PLANNING_YEAR}-07-28"].plot(ax=axes[k,l])
        (supp_regional[c]*1e-3).loc[f"{PLANNING_YEAR}-07-22":f"{PLANNING_YEAR}-07-28"].plot(ax=axes[k,l], color = "orange", alpha = 0.5)
        axes[k,l].set_title(c)
        axes[k,l].set_ylim(0, max_val*1e-3)

        if l ==0:
            axes[k,l].set_ylabel("Withdrawal or supply [GWh]")
        if k==0:
            axes[k,l].set_xlabel("Time")
    axes[0, 4].legend(["Withdrawal (blue)", "Supply (orange)"])
    fig.tight_layout()
    fig.subplots_adjust(hspace=0.15, wspace=0.1)

plot_sufficience_panels(consum_regional, supp_regional)

exports_regional =balance
max_val = balance.max().max()
norm_price =  n.buses_t["marginal_price"][PROV_NAMES].loc[f"{PLANNING_YEAR}-12-22":f"{PLANNING_YEAR}-12-28"]
norm_price/= norm_price.max()
norm_price*= max_val*0.3e-3

def plot_exports_panels(exports_regional, norm_price):
    fig, axes = plt.subplots(6,6, sharex=True, sharey=True, figsize=(20,20))

    for i, c in enumerate(exports_regional.columns):
        k, l = (i+4)//6, (i+4)%6
        exports = (exports_regional[c]*1e-3).loc[f"{PLANNING_YEAR}-12-22":f"{PLANNING_YEAR}-12-28"]
        exports.plot(ax=axes[k,l], color = "orange", alpha = 1)
        if c in norm_price.columns:
            norm_price[c].plot(ax=axes[k,l], color = "green", alpha = 0.5, legend = False, ls ="--")

        axes[k,l].set_title(c)
        # axes[k,l].set_ylim( exports.min(), exports.max())
        axes[k,l].axhline(0, color="black", lw=1)
        if l ==0:
            axes[k,l].set_ylabel("Balance [GWh] \n Norm Price (green)")
        if k==0:
            axes[k,l].set_xlabel("Time")
    axes[0, 4].legend(["Exports"])
    fig.tight_layout()
    fig.subplots_adjust(hspace=0.2, wspace=0.1)

plot_exports_panels(exports_regional, norm_price)

from constants import PROV_NAMES 
ax = n.buses_t["marginal_price"][PROV_NAMES].loc[f"{PLANNING_YEAR}-07-22":f"{PLANNING_YEAR}-07-28"].plot(cmap="viridis",alpha =0.7, logy=True)
ax.set_ylabel("Marginal Price [€/MWh]")
ax.set_xlabel("Time")
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5), title="Province", ncol=2)

<matplotlib.legend.Legend at 0x7f086bc93a10>

rev_and_exp = n.statistics.revenue(bus_carrier="AC", aggregate_time=False).fillna(0).groupby(level=1).sum().T
prod = n.statistics.energy_balance(bus_carrier="AC", aggregate_time=False).fillna(0)
prod = prod[~(prod<=0).all(axis=1)]
prod=prod.clip(lower=1e-5).groupby(level=1).sum().T
hourly_price = rev_and_exp.clip(lower=0)/prod
hourly_price = hourly_price.loc[:,(hourly_price>0).all(axis=0)]
ax = hourly_price.plot(logy=False, cmap = "viridis")
ax.set_ylim([1, hourly_price.max().max()*1.2])
ax.set_ylabel("Revenue/Supply [€/MWh]")
ax.set_xlabel("Time")
ax.legend(loc='center left', bbox_to_anchor=(1, 0.5))

ax2 = (prod/1e6).plot(cmap="viridis", logy=False)
ax2.legend(loc='center left', bbox_to_anchor=(1, 0.5), title="Carrier", ncol=2)
ax2.set_ylabel("Supply [TWh]")
ax2.set_xlabel("Time")
ax2.legend(loc='center left', bbox_to_anchor=(1, 0.5))

<matplotlib.legend.Legend at 0x7f086b9c2c60>

colors_lowered = {k.lower().rstrip():v for k,v in config["plotting"]["tech_colors"].items()}
cap_colors = {k: colors_lowered.get(k.lower().rstrip(), "lightgrey") for k in [c for c in capacity_factors.columns]}

# Apply rolling moving average filter
capacity_factors_smoothed = capacity_factors.rolling(window=24, min_periods=1).mean()

axes = capacity_factors_smoothed.dropna(axis=1, how="all").fillna(0).plot(subplots=True, figsize=(10,12), color= cap_colors)
for ax in axes:
    ax.set_ylim([0,1.1])
    ax.set_yticks([0,0.3, 0.6,1])

COSTS & prices

Price duration curve

n.buses.carrier[n.buses.carrier=="H2"].index.values

array(['Anhui H2', 'Beijing H2', 'Chongqing H2', 'Fujian H2', 'Gansu H2',
       'Guangdong H2', 'Guangxi H2', 'Guizhou H2', 'Hainan H2',
       'Hebei H2', 'Heilongjiang H2', 'Henan H2', 'Hubei H2', 'Hunan H2',
       'InnerMongolia H2', 'Jiangsu H2', 'Jiangxi H2', 'Jilin H2',
       'Liaoning H2', 'Ningxia H2', 'Qinghai H2', 'Shaanxi H2',
       'Shandong H2', 'Shanghai H2', 'Shanxi H2', 'Sichuan H2',
       'Tianjin H2', 'Tibet H2', 'Xinjiang H2', 'Yunnan H2',
       'Zhejiang H2'], dtype=object)

from plot_time_series import plot_price_duration_curve, plot_price_duration_by_node, plot_price_heatmap

plot_price_heatmap(n, "AC", log_values=True)

<Axes: title={'center': 'Heatmap of Log-Transformed Nodal Prices'}, xlabel='Time', ylabel='Nodes'>

plot_price_duration_curve(n, carrier="AC", figsize = (8, 6));

plot_price_duration_by_node(n, carrier="AC", logy=False);

plot_price_duration_by_node(n, carrier="AC");

plot_load_duration_by_node(n, carrier="H2", logy=False, fig_shape=(8, 4), y_lower=0)

<Axes: >

from plot_summary_all import plot_prices
paths = [os.path.join(results_dir, "summary", f"ntwk_{yr}") for yr in snakemake.config["scenario"]["planning_horizons"]]
data_paths = {
        "energy": [os.path.join(p, "energy.csv") for p in paths],
        "costs": [os.path.join(p, "costs.csv") for p in paths],
        "co2_price": [os.path.join(p, "metrics.csv") for p in paths],
        "time_averaged_prices": [os.path.join(p, "time_averaged_prices.csv") for p in paths],
        "weighted_prices": [os.path.join(p, "weighted_prices.csv") for p in paths],
        "co2_balance": [os.path.join(p, "co2_balance.csv") for p in paths],
        "energy_supply": [os.path.join(p, "supply_energy.csv") for p in paths],
        "capacity": [os.path.join(p, "capacities.csv") for p in paths],
    }

fig, ax = plt.subplots()
plot_prices(
    data_paths["time_averaged_prices"],
    config["plotting"],
    fig_name=None,
    ax=ax
)

fig, ax = plt.subplots()
plot_prices(
    data_paths["weighted_prices"],
    config["plotting"],
    fig_name=None,
    ax=ax,
)
# ax.semilogy()

stores_inflow = n.stores_t.p.sum()
h2_stores= stores_inflow.loc[[c for c in stores_inflow.index if c.find("H2 Store")!=-1]]
h2_stores_inflow = h2_stores.where(h2_stores>0,0).sum()
h2_stores_outflow = h2_stores.where(h2_stores<0,0).sum()

STATS

def get_solver_tolerance(config: dict, tol_name = "BarConvTol")->float:
    """get the solver tolerance from the config 

    Args:
        config (dict): the config
        tol_name (str): the name of the tolerance option. Defaults to "BarConvTol"

    Returns:
        float: the value
    """    
    solver_opts = config["solving"]["solver"]["options"]
    return config["solving"]["solver_options"][solver_opts][tol_name]

def find_numerical_zeros(n, config, tolerance_name = "BarConvTol")->list:
    """
    Identify numerical zeros in the network's optimization results.

    This function checks for numerical zeros in the network's optimization results, 
    such as link capacities or weighted prices, based on a specified solver tolerance.

    Args:
        n (pypsa.Network): The PyPSA network object containing optimization results.
        config (dict): Configuration dictionary containing solver options.
        tolerance_name (str): The name of the solver tolerance option to use. Defaults to "BarConvTol".

    Returns:
        list: A list of items (e.g., links or buses) where numerical zeros are detected.
    """
 
    tol = get_solver_tolerance(config, tolerance_name)
    threshold = n.objective*float(tol)
    
    costs = pd.concat([n.statistics.expanded_capex(), n.statistics.opex()],axis=1)
    return costs.fillna(0).sum(axis=1).loc[costs.sum(axis=1) < threshold].index
    
num_zeros = find_numerical_zeros(n, config, tolerance_name = "BarConvTol")

bus_carrier = "AC"
n.loads.carrier = "load"
n.carriers.loc["load", ["nice_name", "color"]] = "Load", "#110d63"
colors = n.carriers.set_index("nice_name").color.where(
    lambda s: s != "", "lightgrey"
)

def rename_index(ds):
    specific = ds.index.map(lambda x: f"{x[1]}\n({x[0]})")
    generic = ds.index.get_level_values("carrier")
    duplicated = generic.duplicated(keep=False)
    index = specific.where(duplicated, generic)
    return ds.set_axis(index)

def plot_static_per_carrier(ds, ax, drop_zero=True):
    if drop_zero:
        ds = ds[ds != 0]
    ds = ds.dropna()
    c = colors[ds.index.get_level_values("carrier")]
    ds = ds.pipe(rename_index)
    label = f"{ds.attrs['name']} [{ds.attrs['unit']}]"
    ds.plot.barh(color=c.values, xlabel=label, ax=ax)
    ax.grid(axis="y")

fig, ax = plt.subplots()
ds = n.statistics.capacity_factor(bus_carrier=bus_carrier).dropna()
plot_static_per_carrier(ds, ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.installed_capacity(bus_carrier=bus_carrier).dropna()
if "Line" in ds.index:
    ds = ds.drop("Line")
ds = ds.drop(("Generator", "Load"), errors="ignore")
ds = ds / 1e3
ds.attrs["unit"] = "GW"
plot_static_per_carrier(ds.abs(), ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.optimal_capacity(bus_carrier=bus_carrier)
if "Line" in ds.index:
    ds = ds.drop("Line")
ds = ds.drop(("Generator", "Load"), errors="ignore")
ds = ds.abs() / 1e3
ds.attrs["unit"] = "GW"
plot_static_per_carrier(ds, ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.capex(bus_carrier=bus_carrier)
plot_static_per_carrier(ds, ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.opex(bus_carrier=bus_carrier)
plot_static_per_carrier(ds, ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.curtailment(bus_carrier=bus_carrier)
plot_static_per_carrier(ds, ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.supply(bus_carrier=bus_carrier)
if "Line" in ds.index:
    ds = ds.drop("Line")
ds = ds / 1e6
ds.attrs["unit"] = "TWh"
plot_static_per_carrier(ds, ax)
plt.show()

if snakemake.config["heat_coupling"]:
    fig, ax = plt.subplots()
    ds = n.statistics.supply(bus_carrier="heat")
    if "Line" in ds.index:
        ds = ds.drop("Line")
    ds = ds / 1e6
    ds.attrs["unit"] = "TWh"
    plot_static_per_carrier(ds, ax)
    plt.show()

fig, ax = plt.subplots()
ds = n.statistics.withdrawal(bus_carrier=bus_carrier)
if "Line" in ds.index:
    ds = ds.drop("Line")
ds = ds / -1e6
ds.attrs["unit"] = "TWh"
plot_static_per_carrier(ds, ax)
plt.show()

fig, ax = plt.subplots()
ds = n.statistics.market_value(bus_carrier=bus_carrier)
plot_static_per_carrier(ds, ax)
plt.show()

def calc_lcoe(n: pypsa.Network, grouper = pypsa.statistics.get_carrier_and_bus_carrier, **kwargs)->pd.DataFrame:
    """calculate the LCOE for the network: (capex+opex)/supply.

    Args:
        n (pypsa.Network): the network for which LCOE is to be calaculated
        grouper (function | list, optional): function to group the data in network.statistics.
                Overwritten if groupby is passed in kwargs.
                Defaults to pypsa.statistics.get_carrier_and_bus_carrier. 
        **kwargs: other arguments to be passed to network.statistics
    Returns:
        pd.DataFrame: The LCOE for the network  with or without brownfield CAPEX. MV and delta

    """    
    if "groupby" in kwargs:
        grouper = kwargs.pop("groupby")

    rev= n.statistics.revenue(groupby=grouper,**kwargs)
    opex = n.statistics.opex(groupby=grouper,**kwargs)
    capex = n.statistics.expanded_capex(groupby=grouper,**kwargs)
    tot_capex = n.statistics.capex(groupby=grouper,**kwargs)
    supply = n.statistics.supply(groupby=grouper,**kwargs)

    profits = pd.concat([opex, capex, tot_capex, rev, supply], axis=1,  keys = ["OPEX", "CAPEX", "CAPEX_wBROWN", "Revenue", "supply"]).fillna(0)
    profits["rev-costs"]=profits.apply(lambda row: row.Revenue-row.CAPEX-row.OPEX, axis=1)
    profits["LCOE"] = profits.apply(lambda row: (row.CAPEX+row.OPEX)/row.supply, axis=1)
    profits["LCOE_wbrownfield"] = profits.apply(lambda row: (row.CAPEX_wBROWN+row.OPEX)/row.supply, axis=1)
    profits["MV"] = profits.apply(lambda row: row.Revenue/row.supply, axis=1)
    profits["profit_pu"]=profits["rev-costs"]/profits.supply
    profits.sort_values("profit_pu", ascending=False, inplace=True)
    
    return profits[profits.supply>0]

rev_costs = calc_lcoe(n, groupby=None)

/p/tmp/ivanra/anaconda/ipykernel_3020255/2397917093.py:25: RuntimeWarning: invalid value encountered in scalar divide
  profits["LCOE"] = profits.apply(lambda row: (row.CAPEX+row.OPEX)/row.supply, axis=1)
/p/tmp/ivanra/anaconda/ipykernel_3020255/2397917093.py:26: RuntimeWarning: invalid value encountered in scalar divide
  profits["LCOE_wbrownfield"] = profits.apply(lambda row: (row.CAPEX_wBROWN+row.OPEX)/row.supply, axis=1)
/p/tmp/ivanra/anaconda/ipykernel_3020255/2397917093.py:27: RuntimeWarning: divide by zero encountered in scalar divide
  profits["MV"] = profits.apply(lambda row: row.Revenue/row.supply, axis=1)

with io.capture_output() as captured:
    s = rev_costs["LCOE"]
    ds.attrs={"name":"LCOE", "unit":"€/MWh"}
    fig, ax = plt.subplots()
    plot_static_per_carrier(ds[~ds.index.isin(num_zeros)], ax)

    ds = rev_costs["profit_pu"]
    ds.attrs={"name":"MV - LCOE", "unit":"€/MWh"}
    fig, ax = plt.subplots()
    plot_static_per_carrier(ds[~ds.index.isin(num_zeros)], ax)

Network topology

irrelevant as don't have lines

n.determine_network_topology()
n.sub_networks["n_branches"] = [
    len(sn.branches()) for sn in n.sub_networks.obj
]
n.sub_networks["n_buses"] = [len(sn.buses()) for sn in n.sub_networks.obj]

CO2 costs

if "emission_prices" in n.meta:
    em_price_gas = float(n.statistics.supply(comps="Generator")["gas"]*n.meta["emission_prices"]["co2"]*0.2/0.43)
    print(f"{em_price_gas:.2e}")
n.global_constraints

2.51e+08

attribute	type	investment_period	carrier_attribute	sense	constant	mu
GlobalConstraint

# costs_s = costs_.T.sum()
# dt = 5
# costs_s.index = costs_s.index.astype(int)
# co2_costs = co2.set_index("Year")["costs bnE"]
# costs_df = pd.DataFrame({'Costs pw bn eur': costs_s, 'CO2 Costs bn eur': co2_costs})
# costs_df["Costs no co2 bn eur"] = costs_df["Costs pw bn eur"] - costs_df["CO2 Costs bn eur"]
# costs_df["discount"]=costs_df.apply(lambda x: 1.02**(x.name-min(costs_df.index)), axis =1)
# discounted_costs = costs_df.apply(lambda x: x/x.discount, axis=1)
# discounted_costs, costs_df.apply(lambda x: x.sum()*5 ), discounted_costs.apply(lambda x: x.sum()*5 )

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[276], line 1
----> 1 costs_s = costs_.T.sum()
      2 dt = 5
      3 costs_s.index = costs_s.index.astype(int)

NameError: name 'costs_' is not defined

# ax = costs_.T.unstack(level=0).T.sort_index().plot(
#     kind='bar', 
#     stacked=True, 
#     figsize=(12, 6), 
#     width=0.8, 
#     legend=True)


# # Customize the legend with a dictionary, including bbox_to_anchor
# legend_kw = {
#     'title': 'Carrier',          # Title of the legend
#     'loc': 'upper left',              # Legend location (usually 'upper left' or 'best')
#     'fontsize': 'small',              # Font size for the legend
#     'frameon': False,                 # Remove the legend's frame
#     'ncol': 2,                        # Number of columns in the legend
#     'bbox_to_anchor': (1.05, 1),      # Adjust the legend's position (x, y)
#     'borderaxespad': 0.5              # Padding between the legend and plot
# }

# # Apply the legend customization
# ax.legend(**legend_kw)

<matplotlib.legend.Legend at 0x7f0e63723c20>

Validation

import seaborn as sns

capex_ = n.statistics.expanded_capex().sum()
installed_capex_ = n.statistics.installed_capex().sum()
opex_ = n.statistics.opex().sum()
costs_ = pd.DataFrame(
    {
        "OPEX": opex_,
        "Installed CAPEX": installed_capex_,
        "CAPEX": capex_,
        "total": capex_ + installed_capex_ + opex_,
    }, index = ["Costs"]
).stack()
revenue_ = n.statistics.revenue(comps="Load", bus_carrier="AC").sum()
costs_.loc[("Revenue", "AC")]=revenue_*-1
all_ = pd.concat({f"{PLANNING_YEAR} budget":costs_}, names=["CO2 control"])
all_.index.names = ["CO2 control", "Cat", "Type"]
all_ = pd.DataFrame(all_, columns = ["Value"])


sns.catplot(all_,x="Cat", y="Value",  hue="Type", col = "CO2 control", dodge=True, kind="bar", alpha =0.8)

<seaborn.axisgrid.FacetGrid at 0x7f087a6d7860>

rev= n.statistics.revenue(groupby=pypsa.statistics.get_carrier_and_bus_carrier)
opex = n.statistics.opex(groupby=pypsa.statistics.get_carrier_and_bus_carrier)
capex = n.statistics.expanded_capex(groupby=pypsa.statistics.get_carrier_and_bus_carrier)
supply = n.statistics.supply(groupby=pypsa.statistics.get_carrier_and_bus_carrier)
index = rev.index.union(opex.index).union(capex.index)
rents = pd.concat([opex, capex, rev, supply], axis=1,  keys = ["OPEX", "CAPEX", "Revenue", "supply"]).fillna(0)
rents["delta"]=rents.apply(lambda row: row.Revenue-row.CAPEX-row.OPEX, axis=1)
rents["delta_pu"]=rents.delta/rents.supply
rents.sort_values("delta_pu", ascending=False, inplace=True)

rents[rents.supply>1e5].droplevel(0).sort_values("delta_pu", ascending=False).plot.bar( y="delta_pu", stacked=True, figsize=(10, 6), ylabel="Rev-Costs Eur/MWh");

EXAMPLES